Use Case:
- Actor Description:
The actor for this app will most likely be someone within the DC
government’s Department of Energy and Environment or Health and Human
Services. The purpose of the visual representation and statistical
modeling would be to assist in the formulation of policy that attempts
to mitigate the occurrences of heat-related health incidences.
- Questions of Interest:
The use of this data frame lends itself to several questions
regarding the correlates that exacerbate extreme heat and the
communities most affected. Thus, our app will assist in understanding of
two specific questions:
What are environmental metrics that are correlated with Urban
Heat Island Effects?
What environmental characteristics of certain census tracts create
larger heat exposure sensitivity indexes? Understanding the
environmental factors would help in identifying policy approaches that
would mitigate the probability of heat-induced health events.
What are recurring sociodemographic trends that are
associated with higher heat exposure sensitivity indexes?
Who are the communities that are at a larger risk of heat-induced
health events? Identifying socio-demographic trends can assist
policymakers in identifying the communities that would best benefit from
policy implementation.
- Choice of approach to data:
Our approach is to be able to map different socio-demographic metrics
and environmental metrics to different Census tracts. This will allow
the user to be able to visualize different metrics and how they
differentiate between census tracts in DC. The modeling will also help
shed insight on the magnitude of these differences. We additionally add
two policy relevant data sets to map including the location and hours of
cooling centers and a random sample of tree data.
Required Packages:
Administrative Packages
shiny (Version 1.7.3): Shiny was used for base layout and skeleton
code of the app as well as functions that assisted in rendering graphs,
maps, models, and data.
thematic (Version 0.1.2.1): Thematic was used to apply a theme for
the app. The theme used was thematic_shiny.
viridis (Version 0.6.2): Viridis was used to create color palettes
for the graphs and maps.
Data Manipulation
tidyverse (Version 1.3.2): Specific tidyverse functions from packages
were used to manipulate the data frame gathered from Open Data DC.
Select and mutate were used heavily to obtain the correct data to work
with throughout the app.
Graphical Analysis
ggplot2 (Version 3.3.6): Also apart of the tidyverse, ggplot2 was
used to render all graphs in the graphing tab. Functions of ggplot were
utilized to successfully add logging, faceting, and smoothing options
for EDA.
Statistical Modeling
shinycssloaders (Version 1.0.0) : shinycssloaders can automatically
load the animations of our Shiny output (e.g. plots, tables). The shiny
output function we used is called withSpinner().
shinydashboard (Version 0.7.2): Shinydashboard was used to easily
create our dashboard for the Regression Analysis tab.
corrplot (Version 0.92) : Corrplot was used to help graphically
display our correlation matrix.
recipes(version 1.0.3) : Recipes was used to assist in creating the
regression equations. This functions lets the variables have arbitrary
roles beyond just predictors and outcomes.
Mapping
leaflet (Version 2.1.1): Leaflet was used to create the mapping tab
of the app. A base level map was created using a shape variable for each
DC census tract. More functions were used to interactively plot tract
metrics, cooling center information, and tree information.
sf (Version 1.0.8): The sf package was used in the business section
to correctly manipulate the shape variable for DC census tracts into the
correct format.
Working Data
tidyverse (Version 1.3.2): The select function of the dplyr package
was used to assist in filtering data based on what input controls were
selected.
Data Source and Structure:
HSEI Data
Heat Sensitivity Exposure Data was gathered from Open Data DC and
contains variables related to socioeconomic, health, and environmental
metrics for DC’s 206 census tracts.
DC Cooling Center Data
When the temperature or heat index in the District reaches 95
degrees, District Government, through the Department of Human Services
(DHS) and the Homeland Security and Emergency Management Agency (HSEMA),
will implement the Heat Emergency Plan and activate cooling centers for
residents to seek relief. We provide locations, name of center, phone
number and hours of cooling centers. Data comes from Open Data DC.
DC Tree Data
DDOT’s Urban Forestry Division (UFD) is the primary steward of
Washington DC’s ~170,000 public trees and has a mission of keeping this
resource healthy, safe, & growing! Trees provide many benefits to
the populations such as improving air and water quality and creates
habitat for local fauna. Importantly trees are also critical for keeping
the city cooler. Data comes from Open Data DC.
